AI

When Jennifer got a job doing research for a nonprofit in 2023, she ran her new professional headshot through a facial recognition program. She wanted to see if the tech would pull up the porn videos she’d made more than 10 years before, when she was in her early 20s. It did in fact return some of that content, and also something alarming that she’d never seen before: one of her old videos, but with someone else’s face on her body. “At first, I thought it was just a different person,” says Jennifer, who is being identified by a pseudonym to protect her privacy.  But then she recognized a distinctly garish background from a video she’d shot around 2013, and she realized: “Somebody used me in a deepfake.” Eerily, the facial recognition tech had identified her because the image still contained some of Jennifer’s features—her cheekbones, her brow, the shape of her chin. “It’s like I’m wearing somebody else’s face like a mask,” she says.  “It’s like I’m wearing somebody else’s face like a mask.” Conversations about sexualized deepfakes—which fall under the umbrella of nonconsensual intimate imagery, or NCII—most often center on the people whose faces are featured doing something they didn’t really do or on bodies that aren’t really theirs. These are often popular celebrities, though over the past few years more people (mostly women and sometimes youths) have been targeted, sparking alarm, fear, and even legislation. But these discussions and societal responses usually are not concerned with the bodies the faces are attached to in these images and videos. As Jennifer, now 37 and a psychotherapist working in New York City, says: “There’s never any discussion about Whose body is this?”  For years, the answer has generally been adult content creators. Deepfakes in fact earned their name back in November 2017, when someone with the Reddit username “deepfakes” uploaded videos showing faces of stars like Scarlett Johansson and Gal Gadot pasted onto porn actors’ bodies. The nonconsensual use of their bodies “happens all the time” in deepfakes, says Corey Silverstein, an attorney specializing in the adult industry.  But more recently, as generative AI has improved, and as “nudify” apps have begun to proliferate, the issue has grown far more complicated—and, arguably, more dangerous for creators’ futures.  Porn actors’ bodies aren’t necessarily being taken directly from sexual images and videos anymore, or at least not in an identifiable way. Instead, they are inevitably being used as training data to inform how new AI-generated bodies look, move, and perform. This threatens the livelihood and rights of porn actors as their work is used to train AI nudes that in turn could take away their business. And that’s not all: Advancements in AI have also made it possible for people to wholly re-create these performers’ likenesses without their consent, and the AI copycats may do things the performers wouldn’t do in real life. This could mean their digital doubles are participating in certain sex acts that they haven’t agreed to do, or even that they’re perpetrating scams against fans.  Adult content creators are already marginalized by a society that largely fails to protect their safety and rights, and these developments put them in an even more vulnerable position. After Jennifer found the deepfake featuring her body, she posted on social media about the psychological effects: “I’ve never seen anyone ask whether that might be traumatic for the person whose body was used without consent too. IT IS!” Several other creators I spoke with shared the mental toll that comes with knowing their bodies have been used nonconsensually, as well as the fear that they’ll suffer financially as other people pirate their work. Silverstein says he hears from adult actors every day who “are concerned that their content is being exploited via AI, and they’re trying to figure out how to protect it.”  One law professor and expert in violence against women calls these creators the “forgotten victims” of NCII deepfakes. And several of the people I spoke with worry that as the US develops a legal framework to combat nonconsensual sexual content online, adult actors are only at risk of further injury; instead of helping them, the crackdown on deepfakes may provide a loophole through which their content and careers could be stripped from the internet altogether. How deepfakes cause “embodied harms” During his preteen years in the 1970s, Spike Irons, now a porn actor and president of the adult content platform XChatFans, was “in love” with Farrah Fawcett. Though Fawcett did not pose nude, Jones managed to get his hands on what looked like pictures of her naked. “People were cutting out faces and pasting them on bodies,” Irons says. “Deepfakes, before AI, had been going around for quite a while. They just weren’t as prolific.” The early public internet was rife with websites capitalizing on the idea that you could use technology to “see” celebrities naked. “People would just use Microsoft Paint,” says Silverstein, the attorney. It was a simple way to mash up celebrities’ faces with porn.  People later used software like Adobe After Effects or FakeApp, which was designed to swap two individuals’ faces in images or videos. None of these programs required serious expertise to alter content, so there was a low barrier to entry. That, plus the wealth of porn performers’ videos online, helped make face-swap deepfakes that used real bodies prevalent by the 2010s. When, later in the decade, deepfakes of Gal Gadot and Emma Watson caused something of a broader panic, their faces were allegedly swapped onto the bodies of the porn actors Pepper XO and Mary Moody, respectively. But it wasn’t just high-profile actors like them whose bodies were being used. Jennifer was “a very minor performer,” she says. “If it happened to me, I feel like it could happen to anybody who’s shot porn.” Since he started his practice in 2006, Silverstein says, “numerous clients” have reached out to report “This is my body on so-and-so.”  Both people whose faces

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. The shock of seeing your body used in deepfake porn When Jennifer got a research job in 2023, she ran her new professional headshot through a facial recognition program. She wanted to see whether it would pull up the porn videos she’d made more than a decade earlier. It did, but it also surfaced something she’d never seen before: one of her old videos, now featuring someone else’s face on her body. Conversations about sexualized deepfakes usually focus on the people whose faces are inserted into explicit content without consent. But another group often gets ignored: the people whose bodies those faces are attached to. Adult content creators say AI systems are training on their work, cloning their likenesses, and generating explicit content they never agreed to make, all with little legal protection or control.  Read the full story on the threat to their rights, livelihoods, and ownership of their own bodies. —Jessica Klein This story is part of our The Big Story series, the home for MIT Technology Review’s most important, ambitious reporting. You can read the rest here.  AI chatbots are giving out people’s real phone numbers Generative AI is exposing people’s personal contact information—and there’s no easy way to stop it. A software developer started receiving WhatsApp messages asking for help after Gemini surfaced his number. A university researcher got the chatbot to reveal a colleague’s private cell number. A Reddit user says Gemini sent a stream of callers looking for lawyers to his phone. Experts believe these privacy lapses stem from personally identifiable information in AI training data. Chatbots may now be making that information dramatically easier to find. Find out why these breaches are growing—and why there’s little that victims can do to stop them. —Eileen Guo The Tesla Semi could be a big deal for electric trucking Nearly a decade after Elon Musk first unveiled the Tesla Semi, the electric truck is finally rolling off the production line. It could be a breakout moment for battery-powered freight. Semitrucks produce an outsized share of road transport pollution, while electric alternatives have struggled with high prices, limited range, and charging challenges. Tesla is betting the Semi can overcome those problems. The truck reportedly travels up to 480 miles on a single charge and costs far less than many competing electric models. Here’s how the Tesla Semi could give electric trucking a vital boost. —Casey Crownhart This article is from The Spark, MIT Technology Review’s weekly climate newsletter. To receive it in your inbox every Wednesday, sign up here. The must-reads I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology. 1 The US has approved Nvidia chip sales to 10 Chinese firmsAlibaba, Tencent, and ByteDance are among those cleared to buy H200 chips. (Reuters $)+ The US will receive 25% of the revenue from the sales. (Engadget)+ But Beijing wants domestic firms to prioritize homegrown chips. (Nikkei Asia)+ Nvidia CEO Jensen Huang is in China with a White House delegation. (CNBC) 2 Beijing’s push for AI independence is weakening US leverageIt’s allowing China to resist pressure during the Beijing talks. (NYT $)+ The country has made a big bet on open-source. (MIT Technology Review)+ Here’s what’s at stake for tech at the Trump-Xi meeting. (Rest of World) 3 AI is “rotting the brains” of developersThey’re losing their previous abilities to do their jobs. (404 Media)+ A populist backlash is building against AI. (MIT Technology Review)+ It’s time to reset our expectations about AI. (MIT Technology Review) 4 Sam Altman has over $2 billion in companies that have dealt with OpenAIThe ties have triggered accusations of conflicts of interest. (The Times $)+ The GOP is scrutinizing Altman’s business dealings. (WSJ $) 5 Andreessen Horowitz has become the top political donor in the US A16z contributed $115.5 million to the midterm elections. (NYT)+ AI lobbying has reached a fever pitch. (NYT $)6 Microsoft feared being too dependent on OpenAI  CEO Satya Nadella was worried about OpenAI supplanting his company. (CNBC)+ Microsoft is eyeing startup deals for life after OpenAI. (Reuters $) 7 AI systems are forecasting wars and regime collapseOne estimates a 20% chance of regime change in Iran by 2026. (Economist $)+ AI has turned the Iran conflict into theater. (MIT Technology Review) 8 Anthropic says a model behaved badly due to training on dystopian sci-fiTraining on more positive stories could help. (Ars Technica) 9 Data centers now consume 6% of the electricity in the US and UKAI’s global energy consumption is up 15% globally in two years. (Guardian) 10 NASA has rescued Curiosity after its drill got stuck on MarsThe agency has just revealed how it freed the rover. (Wired $)  Quote of the day “Musk loves to be glazed, and this person is the doughnut factory.” —Joan Donovan, assistant professor of journalism and emerging media studies at Boston University, tells the Washington Post how Elon Musk has consistently amplified one anonymous X account. One More Thing YOSHI SODEOKA Inside the messy ethics of making war with machines In a near-future war—one that might begin tomorrow—a sniper’s computer vision system flags a potential target. Just over the horizon, a chatbot advises a commander to order an artillery strike. In both cases, an AI system recommends pulling the trigger while a human still has the final say. But how much of the decision is really theirs? When, if ever, is it ethical for that decision to kill? And who’s to blame when something goes wrong? This is how AI is reshaping decision-making on the battlefield. —Arthur Holland Michel We can still have nice things A place for comfort, fun, and distraction to brighten up your day. (Got any ideas? Drop me a line.) + The secrets behind how Shazam works have been revealed.+ For the first time in a decade, a rare “Cloud Jaguar” was caught on camera.+ Explore

When generative AI first moved from research labs into real-world business applications, enterprises made a tacit bargain: “Capability now, control later.” Feed your proprietary data into third-party AI models, and you will get powerful results. But your data passes through systems you do not own, under governance you do not set. The protections you rely on are only as durable as the provider’s next policy update. Now, with generative AI established in everyday business operations and sophisticated new agentic AI systems advancing every day, companies are reevaluating the terms of that deal. “Data is really a new currency; it’s the IP for many companies,” says Kevin Dallas, CEO of EDB, echoing a recurrent anxiety from customers. “The big concern is, if you’re deploying an AI-infused application with a cloud-based large language model, are you losing your IP? Are you losing your competitive position?” DOWNLOAD THE REPORT That question is now fueling a movement toward reclaiming both the data and AI systems that have rapidly become part of core business infrastructure. AI and data sovereignty, which refers to breaking dependence on centralized providers and establishing genuine control over models and data estates, it is an urgent priority for many companies, says Dallas, citing internal EDB data: “70% of global executives believe they need a sovereign data and AI platform to be successful.” The idea of AI sovereignty is becoming a global policy conversation. NVIDIA CEO Jensen Huang recently spoke about the need for such a shift at the World Economic Forum’s annual meeting at Davos in January 2026: “I really believe that every country should get involved to build AI infrastructure, build your own AI, take advantage of your fundamental natural resource—which is your language and culture—develop your AI, continue to refine it, and have your national intelligence be part of your ecosystem.” This report explores how enterprises are pursuing sovereignty over their models and data estates in an era of rapid AI adoption. Drawing on a survey conducted by EDB of more than 2,050 senior executives and a series of interviews with industry experts, the research confirms that the sovereignty movement on the enterprise level is already well underway. Download the report. This content was produced by Insights, the custom content arm of MIT Technology Review. It was not written by MIT Technology Review’s editorial staff. It was researched, designed, and written by human writers, editors, analysts, and illustrators. This includes the writing of surveys and collection of data for surveys. AI tools that may have been used were limited to secondary production processes that passed thorough human review.

Financial services companies have unique needs when it comes to business AI. They operate in one of the most highly regulated sectors while responding to external events that are updated by the second. As a result, the success of agentic AI in financial services depends less on the sophistication of the system and more on the quality, security, and accessibility of the data it relies on.  “It all starts with the data,” says Steve Mayzak, global managing director of Search AI at Elastic. Agentic AI—systems that can independently plan and take actions to complete tasks, rather than simply generate responses—holds enormous potential for financial services due to its ability to incorporate real-time data and optimize complex workflows. Gartner has found that more than half of financial services teams have already implemented or plan to implement agentic AI.  However, introducing autonomous AI into any organization magnifies both the strengths and weaknesses of the underlying data it uses. To deploy agentic AI with speed, confidence, and control, financial services companies must first be able to search, secure, and contextualize their data at scale. “Agentic AI amplifies the weakest link in the chain: data availability and quality,” says Mayzak. “And your systems are only as good as their weakest link.” Financial services companies, therefore, require a trusted and centralized data store that is easy to access, dependable, and can be managed at scale. The high stakes of quality information Regulation in the financial services sector requires a high degree of accountability for all data tools. As Mayzak says, “You can’t just stop at explaining where the data came from and what it was transformed into: ‘Here’s the data that went in, and this is what came out.’ You need an auditable and governable way to explain what information the model found and the logic of why that data was right for the next step.” That is, you need to be able to see, understand, and describe the underlying processes. At the same time, financial services companies require speed and accuracy in order to meet customer expectations and stay ahead of competition. Markets are continually shifting, and risks and opportunities move along with them. If an AI model can parse natural language (unstructured data) from complex sources—in addition to structured data in spreadsheets that are easier to analyze—this gives users more relevant information.  In this environment, there is no tolerance for error, including the hallucinations that plagued early AI efforts. Agentic AI systems depend on rapid access to high-quality, well-governed data that is secure and accessible. In financial services, that data spans transactions, customer interactions, risk signals, policies, and historical context. The task of preparing that data for AI should not be underestimated. “Natural language is way more messy than structured data, and that makes the process of organizing and cleaning it up that much more important and also that much harder,” says Mayzak. The data must be well indexed and consolidated across different locations, not locked in the silos of separate systems across the organization. Otherwise, AI agents lag, provide inconsistent answers, and produce decisions that are harder to trace and explain, undermining confidence among regulators, customers, and internal stakeholders.  As Mayzak says, “There are many different ways to describe how to execute a trade at a bank. In an agent-powered world, we need those descriptions to be deterministic—to give the same results every time. Yet we’re building on powerful but non-deterministic models. That’s incredibly tricky, but not impossible.” For a financial services firm, managing this can be very challenging. A Forrester study found that 57% of financial organizations are still developing the necessary internal capabilities to fully leverage agentic AI. “The data exists in many different formats, created over the course of a bank’s history,” says Mayzak. “Take any bank that’s been around for 50 years: They might have 60 different types of PDFs for the exact same thing. And at the same time, we want the output of these systems to be 100% accurate. In many cases, there is no ‘good enough’.” That is, companies need to do it right, and the first time. Searching and securing results  An effective search platform is key to solving the problem of fragmented, poorly indexed, inaccessible data. Financial services companies that can readily sift through both their structured and unstructured data, keep it secure, and apply it in the right context will get the most value from agentic AI. This often requires designing AI systems with data access and utility in mind so they can work faster and yield more accurate results, as well as reduce risk. “Search is the foundational technology that makes AI accurate and grounded in real data,” Mayzak says. “Search platforms have become the authoritative context and memory stores that will power this AI revolution.” Once in place, these AI-enhanced searches and autonomous systems can serve financial services companies for a range of purposes. When monitoring client exposure, agentic AI can continuously scan transactions, market signals, and external data to detect emerging risks; platforms can then automatically flag or escalate issues in real time. In trade monitoring, AI agents can review trade workflows, identify discrepancies across different formats, and resolve exceptions step by step with minimal human intervention. In regulatory reporting, AI can gather data from across systems, generate required reports, and track how each output was produced. These applications of AI save time while supporting audit and compliance needs by being traceable and explainable. Although such capabilities already exist, they are often manual, fragmented, and difficult to scale. Agentic AI allows financial organizations to move toward more automated, efficient, and scalable processes while maintaining the accuracy and transparency required in their highly regulated environment. As Mayzak says, “It’s not that different from how humans operate today, just done at a much faster pace and at scale.”  Building an agentic AI ecosystem Launching agentic AI can be daunting, especially if other AI ventures have stalled internally. Mayzak’s recommendation is to choose a manageable use case and allow it to

The mouse pointer has sat at the center of personal computing for more than half a century. It tracks cursor position. It registers clicks. Beyond that, it does almost nothing. Google DeepMind researchers outlined a set of experimental principles and demos for an AI-enabled pointer that goes considerably further: one that understands not just where you are pointing, but what you are pointing at and why it matters. The system is powered by Gemini and is currently in the experimental stage. Two demos are live in Google AI Studio today: one for editing an image and one for finding places on a map, both operable by pointing and speaking. A deeper integration called Magic Pointer is also rolling out inside Chrome, and a further integration is planned for Googlebook, Google’s new line of Gemini-powered laptops announced this week. https://deepmind.google/blog/ai-pointer/ What DeepMind is Targeting The frustration DeepMind researchers are addressing is a familiar one for anyone who has tried to use an AI assistant while already in the middle of work. Because a typical AI tool lives in its own window, users need to drag their world into it. The research team wants the opposite — intuitive AI that meets users across all the tools they use, without interrupting their flow. In practice, today’s AI workflow often looks like this: you are working inside a document or a browser tab, you spot something you want to ask about, you switch to a chat interface, you re-describe what you were looking at, you run the query, and you paste the result back. This maps to a concrete technical gap: current LLM interfaces are largely text-in, text-out. They have no awareness of the screen state around them. The AI-enabled pointer is an attempt to close that gap by giving the model real-time visual and semantic context derived from cursor position and hover state — without requiring users to manually serialize that context into a written prompt. Four interaction principles DeepMind researchers have developed four principles that together shift the hard work of conveying context and intent from the user to the computer, replacing text-heavy prompts with simpler, more intuitive interactions. The first is Maintain the flow. AI capabilities should work across all apps, not force users into ‘AI detours’ between them. The prototype AI-enabled pointer is available wherever the user is working. For example, they could point at a PDF and request a bullet-point summary to paste directly into an email, hover over a table of statistics and request a pie chart version, or highlight a recipe and ask for all the ingredients doubled. This is a direct architectural stance: instead of building AI assistance as a sidecar application, the capability lives at the pointer level and is present in whichever tool the user is already working in. The second is Show and tell. Current AI models demand precise instructions. To get a good response, a user has to write a detailed prompt. An AI-enabled pointer would streamline this process by smoothly capturing the visual and semantic context around the pointer, letting the computer ‘see’ and understand what’s important to the user. In the experimental system, just point, and the AI knows exactly which word, paragraph, part of an image, or code block the user needs help with. From a technical standpoint, this means the system treats cursor hover state and the surrounding UI content as structured model inputs — comparable to how multimodal models process image and text together, except here the visual region is dynamically cropped and contextualized in real time around a moving cursor. The third is Embrace the power of ‘This’ and ‘That‘. In everyday interactions with each other, humans rarely speak in long, detailed paragraphs. We might say, ‘Fix this’, ‘Move that here’, or ‘What does this mean?’ — while relying on physical gestures and our shared context to fill in any gaps in understanding. An AI system that understands this combination of context, pointing and speech would allow users to make complex requests in natural shorthand, no fiddly prompting required. The name of the principle is deliberate: deictic language (words like ‘this’ and ‘that’ that depend on physical reference to carry meaning) is how humans naturally communicate when they can point at something. The AI-enabled pointer is designed to handle exactly that class of instruction without needing the user to spell out what “this” refers to. The fourth is Turn pixels into actionable entities. For decades, computers have only tracked where we are pointing. AI can now also understand what the user is pointing at. This transforms pixels into structured entities, such as places, dates, and objects, that users can interact with instantly. A photo of a scribbled note becomes an interactive to-do list; a paused frame in a travel video becomes a booking link for that cool-looking restaurant. For ML engineers, this is the most technically substantive of the four principles. It describes an entity extraction step that happens at inference time on whatever visual content is under the cursor — converting raw pixel regions into typed, actionable objects rather than leaving them as unstructured screen content. Where it is going Google DeepMind is now integrating these principles to reimagine pointing in Chrome and the new Googlebook laptop experience. Starting now, instead of writing a complex prompt, users can use their pointer to ask Gemini in Chrome about the part of the webpage they care about. For example, selecting a few products on a page and asking to compare them, or pointing to where they want to visualize a new couch in their living room. Key Takeaways Google DeepMind introduces experimental demos of an AI-enabled mouse pointer powered by Gemini that captures visual and semantic context around the cursor — no manual prompting required. The system is built on four principles: Maintain the flow, Show and tell, Embrace the power of “This” and “That”, and Turn pixels into actionable entities. “Turn pixels into actionable entities” is the key technical idea — the pointer converts on-screen

Most AI systems today work in turns. You type or speak, the model waits, processes your input, and then responds. That’s the entire interaction loop. Thinking Machines Lab, an AI research lab, is arguing that this model of interaction is a fundamental bottleneck. Thinking Machines Lab team introduced a research preview of a new class of system they call interaction models to address it. The main idea for their research is interactivity should be native to the model itself, not bolted on as an afterthought. What’s Wrong with Turn-Based AI If you’ve built anything with a language model or voice API, you’ve worked around the limitations of turn-based interaction. The model has no awareness of what’s happening while you’re still typing or speaking. It can’t see you pause mid-sentence, notice your camera feed, or react to something visual in real time. While the model is generating, it’s equally blind — perception freezes until it finishes or gets interrupted. This creates a narrow channel for human-AI collaboration that limits how much of a person’s knowledge, intent, and judgment can reach the model, and how much of the model’s work can be understood. To work around this, most real-time AI systems use a harness — a collection of separate components stitched together to simulate responsiveness. A common example is voice-activity detection (VAD), which predicts when a user has finished speaking so a turn-based model knows when to start generating. This harness is made out of components that are meaningfully less intelligent than the model itself, and it precludes capabilities like proactive visual reactions, speaking while listening, or responding to cues that are never explicitly stated aloud. Thinking Machines Lab’s argument is a version of the ‘bitter lesson’ in machine learning: hand-crafted systems will eventually be outpaced by scaling general capabilities. For interactivity to scale with intelligence, it must be part of the model itself. With this approach, scaling a model makes it smarter and a better collaborator. https://thinkingmachines.ai/blog/interaction-models/ The Architecture: Multi-Stream, Micro-Turn Design The system has two components working in parallel: an interaction model that maintains constant real-time exchange with the user, and a background model that handles deeper reasoning tasks asynchronously. The interaction model is always on — continuously taking in audio, video, and text and producing responses in real time. When a task requires sustained reasoning (tool use, web search, longer-horizon planning), it delegates to the background model by sending a rich context package containing the full conversation — not a standalone query. Results stream back as the background model produces them, and the interaction model interleaves those updates into the conversation at a moment appropriate to what the user is currently doing, rather than as an abrupt context switch. Both models share their context throughout. Think of it like one person who keeps you engaged in conversation while a colleague in the background looks something up and passes notes forward in real time. The key architectural decision enabling this is time-aligned micro-turns. The interaction model continuously interleaves the processing of 200ms worth of input with the generation of 200ms worth of output. Rather than consuming a complete user turn and generating a complete response, both input and output are treated as streams processed in 200ms chunks. This is what allows the model to speak while listening, react to visual cues without being prompted verbally, handle true simultaneous speech, and make tool calls and browse the web while the conversation is still in progress — weaving results back in as they arrive. Encoder-free early fusion is the specific design choice that makes multimodal processing work at this cadence. Rather than routing audio and video through large, separate pretrained encoders (like a Whisper-style ASR model or a standalone TTS decoder), the architecture uses minimal pre-processing. Audio signals are ingested as dMel and transformed via a lightweight embedding layer. Video frames are split into 40×40 patches encoded by an hMLP. Audio output uses a flow head for decoding. All components are co-trained from scratch together with the transformer — there is no separately pretrained encoder or decoder at any stage. On the inference side, the 200ms chunk design creates engineering challenges. Existing LLM inference libraries aren’t optimized for frequent small prefills — they carry significant per-turn overhead. Thinking Machines implemented streaming sessions, where the client sends each 200ms chunk as a separate request while the inference server appends chunks into a persistent sequence in GPU memory, avoiding repeated memory reallocations and metadata computations. They’ve upstreamed a version of this to SGLang, the open-source inference framework. Additionally, they use a gather+gemv strategy for MoE kernels instead of standard grouped gemm, following prior work from PyTorch and Cursor, to optimize for the latency-sensitive shapes required by bidirectional serving. https://thinkingmachines.ai/blog/interaction-models/ Benchmarks: Where It Stands The model, named TML-Interaction-Small, is a 276B parameter Mixture-of-Experts (MoE) with 12B active parameters. The benchmark table distinguishes between Instant models (no extended reasoning) and Thinking models (with reasoning). TML-Interaction-Small is an Instant model. Among all Instant models in the comparison, it achieves the highest score on Audio MultiChallenge APR at 43.4% — above GPT-realtime-2.0 (minimal) at 37.6%, GPT-realtime-1.5 at 34.7%, and Gemini-3.1-flash-live-preview (minimal) at 26.8%. The Thinking models, GPT-realtime-2.0 (xhigh) at 48.5% and Gemini-3.1-flash-live (high) at 36.1%, use extended reasoning to achieve their scores. On FD-bench v1.5, which measures interaction quality across user interruption, backchanneling, talking-to-others, and background speech scenarios, TML-Interaction-Small scores 77.8 average quality — compared to 54.3 for Gemini-3.1-flash-live (minimal), 48.3 for GPT-realtime-1.5, and 47.8 for GPT-realtime-2.0 (xhigh). On FD-bench v1 turn-taking latency, the model responds in 0.40 seconds — compared to 0.57s for Gemini, 0.59s for GPT-realtime-1.5, and 1.18s for GPT-realtime-2.0 (minimal). On FD-bench v3, which evaluates response quality and tool use (audio + tools combined), TML-Interaction-Small (with background agent enabled) scores 82.8% Response Quality / 68.0% Pass@1 — the highest in the comparison table. https://thinkingmachines.ai/blog/interaction-models/ Thinking Machines research team also introduced new internal benchmarks targeting capabilities that no existing model handles: TimeSpeak — Tests whether the model initiates speech at user-specified times with correct content.

Most

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. A plan to make drugs in orbit is going commercial A startup called Varda Space Industries is betting that the future of pharmaceuticals lies in orbit. The company has signed a deal with United Therapeutics to test whether drugs crystallize differently in microgravity, potentially creating improved versions with new properties. The idea sounds futuristic, but falling launch costs and reusable rockets are making space-based manufacturing seem increasingly plausible. Varda says the partnership could mark an important step toward building products in orbit for use back on Earth. Discover how space could become the next frontier for drug development. —Antonio Regalado MIT Technology Review Narrated: NASA is building the first nuclear reactor-powered interplanetary spacecraft. How will it work? Just before Artemis II began its historic slingshot around the moon, NASA revealed an even grander space travel plan. By the end of 2028, the agency aims to fly a nuclear reactor-powered interplanetary spacecraft to Mars. A successful mission would herald a new era in spaceflight—and might just give the US the edge in the race against China. But the project remains shrouded in mystery. MIT Technology Review picked the brains of nuclear power and propulsion experts to find out how the nuclear-powered spacecraft might work. —Robin George Andrews This is our latest story to be turned into an MIT Technology Review Narrated podcast, which we publish each week on Spotify and Apple Podcasts. Just navigate to MIT Technology Review Narrated on either platform, and follow us to get all our new content as it’s released. The must-reads I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology. 1 Sam Altman claims Elon Musk tried to seize control of OpenAIAltman said Musk initially wanted 90% of the equity. (AFP)+ And that control should go to his children when he dies. (BBC)+ Altman also accused Musk of twice trying to end its non-profit status. (NPR)+ Musk’s motivations for the suit are under scrutiny. (MIT Technology Review) 2 Google and SpaceX are in talks to launch data centers into orbitSpaceX could join Suncatcher, Google’s orbital data center project. (WSJ $)+ The project’s first launch is slated for early 2027. (Guardian)+ Anthropic and SpaceX have also discussed orbital data centers. (Wired $)+ But there are a few hurdles to overcome. (MIT Technology Review)  3 Jensen Huang has joined Donald Trump’s high-stakes mission to ChinaNvidia is lobbying to sell its AI chips in the country. (Bloomberg $)+ Elon Musk and Tim Cook are also on the trip. (CNBC)+ But a tech rivalry and distrust have sapped hopes for big deals. (Reuters $) 4 ICE agents have a list of 20 million people on their iPhones, thanks to PalantirAn ICE official said Palantir is speeding up raids and arrests. (404 Media)+ ICE has also used facial recognition and Paragon spyware. (TechCrunch) 5 Defense tech firm Anduril just doubled its valuation to over $60 billionIn a $5 billion funding round led by Thrive Capital and a16z. (FT $)Anduril, which makes AI-backed weapons, may go public next year. (NYT $) 6 Meta employees are protesting computer-tracking at workFlyers posted at offices are urging staff to oppose the program. (Reuters $)+ Meta plans to track workers’ clicks and keystrokes to train AI. (CNBC) 7 OpenAI is facing another wrongful death lawsuit over ChatGPT medical adviceThe chatbot’s tips allegedly led to a teenager’s overdose. (Ars Technica) 8 The Canvas learning platform has paid hackers to delete stolen student dataIt caved to ransomware demands after the biggest-ever edtech breach. (BBC) 9 Scientific researchers are thinking twice about using AIDue to price hikes, usage limitations, and unreliable outputs. (Nature) 10 The latest AI compute solution? Putting data centers in your homeHardware hosts get subsidized electricity and internet. (Ars Technica) Quote of the day “Mr Musk did try to kill it.” —Sam Altman claims that Elon Musk tried to destroy rather than protect OpenAI’s non-profit operations, the Guardian reports. One More Thing YOSHI SODEOKA Why does AI hallucinate? Chatbot fails are now a familiar meme. Meta’s short-lived scientific chatbot generated wiki articles about the history of bears in space. Lawyers have submitted court documents filled with legal citations fabricated by ChatGPT. Air Canada was ordered to honor a refund policy invented by its customer service chatbot. This tendency to make things up—known as hallucination—is one of the biggest obstacles holding chatbots back from more widespread adoption. Here’s why they do it—and why we still can’t fix it. —Will Douglas Heaven This story is part of MIT Technology Review Explains, our series untangling the complex, messy world of technology to help you understand what’s coming next. You can read more from the series here.  We can still have nice things A place for comfort, fun, and distraction to brighten up your day. (Got any ideas? Drop me a line.) + A historian has unearthed the etymology of every single dinosaur name.+ Humus on the moon is getting closer to reality after scientists grew chickpeas in lunar soil.+ Witness the patience of a master paper artist in this gallery of intricate, handmade sculptures.+ Want to tell the time alphabetically? Me neither, but this cursed clock is an intriguing reason to try.